Monitoring and control of refrigeration equipment

Abstract

A system generates alerts based on sensor data obtained from equipment, for example, refrigeration equipment. Examples of equipment instances include refrigeration equipment, heating equipment, air conditioning equipment, and so on. The system accesses a model, for example, a machine learning model trained to predict an alert threshold for generation of alerts. The system modifies the alert threshold value for an equipment instance based on the output of the machine learning model. The system uses the modified alert threshold value for generating alerts. The system may track the number of times the alert threshold was adjusted. If the number of times the alert threshold value is modified exceeds a predetermined fault threshold value, the system determines that the equipment is faulty.

What is claimed is: 
     
       1. A computer implemented method for monitoring alerts generated by refrigeration equipment, the method comprising:
 receiving sensor data generated by sensors mounted on a refrigeration equipment instance, wherein the refrigeration equipment instance is a particular installation of a refrigeration equipment; 
 monitoring a rate of generation of alerts by the refrigeration equipment instance, an alert indicating a request for attention by the refrigeration equipment instance, the alert generated responsive to a sensor data exceeding an alert threshold value; 
 repeating one or more times:
 determining that the rate of generation of alerts by the refrigeration equipment exceeds a rate threshold value; 
 modifying the alert threshold value, wherein the modified alert threshold value is expected to reduce the rate of generation of alerts by the refrigeration equipment; and 
 using the modified alert threshold value for generating alerts for the refrigeration equipment instance; 
 
 determining, for the refrigeration equipment instance, a number of times the alert threshold value was modified; and 
 responsive to the number of times the alert threshold value was modified exceeding a fault threshold value, determining that the refrigeration equipment instance is faulty. 
 
     
     
       2. The method of  claim 1 , further comprising:
 determining the modified threshold value by executing a machine learning based model. 
 
     
     
       3. The method of  claim 1 , wherein the modified alert threshold value is determined based on factors including a number of alerts received during a time interval. 
     
     
       4. The method of  claim 1 , wherein the modified alert threshold value is determined based on factors including a measure of an indication that an action was taken to adjust the refrigeration equipment instance in response to an alert sent. 
     
     
       5. The method of  claim 4 , wherein the measure of the indication that an action was taken to adjust the refrigeration equipment instance in response to an alert sent is determined based on a change in sensor data monitored responsive to sending the alert. 
     
     
       6. The method of  claim 1 , wherein the modified alert threshold value is determined based on factors including an expected normal performance of the refrigeration equipment instance. 
     
     
       7. The method of  claim 1 , wherein modifying the alert threshold value comprises determining a first threshold value for a first time interval during each day and a second threshold value for a second time interval during each day. 
     
     
       8. The method of  claim 1 , wherein modifying the alert threshold value comprises determining a first threshold value for a first set of refrigeration equipment instances and a second threshold value for a second set of refrigeration equipment instances. 
     
     
       9. A non-transitory computer-readable storage medium storing computer program instructions executable by one or more electronic processors to perform operations comprising:
 receiving sensor data generated by sensors mounted on a refrigeration equipment instance, wherein the refrigeration equipment instance is a particular installation of a refrigeration equipment; 
 monitoring a rate of generation of alerts by the refrigeration equipment instance, an alert indicating a request for attention by the refrigeration equipment instance, the alert generated responsive to a sensor data exceeding an alert threshold value; 
 repeating one or more times:
 determining that the rate of generation of alerts by the refrigeration equipment exceeds a rate threshold value; 
 modifying the alert threshold value, wherein the modified alert value is expected to reduce the rate of generation of alerts by the refrigeration equipment; and 
 using the modified alert threshold value for generating alerts for the refrigeration equipment instance; 
 
 determining, for the refrigeration equipment instance, a number of times the alert threshold value was modified; and 
 responsive to the number of times the alert threshold value was modified exceeding a fault threshold value, determining that the refrigeration equipment instance is faulty. 
 
     
     
       10. The non-transitory computer-readable storage medium of  claim 9 , wherein the instructions further cause the one or more electronic processors to perform steps comprising:
 determining the modified alert threshold value by executing a machine learning based model. 
 
     
     
       11. The non-transitory computer-readable storage medium of  claim 9 , wherein the modified alert threshold value is determined based on factors including a number of alerts received during a time interval. 
     
     
       12. The non-transitory computer-readable storage medium of  claim 9 , wherein the modified alert threshold value is determined based on factors including a measure of an indication that an action was taken to adjust the refrigeration equipment instance in response to an alert sent. 
     
     
       13. The non-transitory computer-readable storage medium of  claim 12 , wherein the measure of the indication that an action was taken to adjust the refrigeration equipment instance in response to an alert sent is determined based on a change in sensor data monitored responsive to sending the alert. 
     
     
       14. The non-transitory computer-readable storage medium of  claim 9 , wherein the modified alert threshold value is determined based on factors including an expected normal performance of the refrigeration equipment instance. 
     
     
       15. The non-transitory computer-readable storage medium of  claim 9 , wherein modifying the alert threshold value comprises determining a first threshold value for a first time interval during each day and a second threshold value for a second time interval during each day. 
     
     
       16. The non-transitory computer-readable storage medium of  claim 9 , wherein modifying the alert threshold value comprises determining a first threshold value for a first set of refrigeration equipment instances and a second threshold value for a second set of refrigeration equipment instances. 
     
     
       17. A computer system comprising:
 one or more electronic processors; and 
 a non-transitory computer-readable storage medium storing computer program instructions executable by the one or more electronic processors to perform operations comprising:
 receiving sensor data generated by sensors mounted on a refrigeration equipment instance, wherein the refrigeration equipment instance is a particular installation of a refrigeration equipment; 
 monitoring a rate of generation of alerts by the refrigeration equipment instance, an alert indicating a request for attention by the refrigeration equipment instance, the alert generated responsive to a sensor data exceeding an alert threshold value; 
 repeating one or more times:
 determining that the rate of generation of alerts by the refrigeration equipment exceeds a rate threshold value; 
 modifying the alert threshold value, wherein the modified alert value is expected to reduce the rate of generation of alerts by the refrigeration equipment; and 
 using the modified alert threshold value for generating alerts for the refrigeration equipment instance; 
 
 determining, for the refrigeration equipment instance, a number of times the alert threshold value was modified; and 
 responsive to the number of times the alert threshold value was modified exceeding a fault threshold value, determining that the refrigeration equipment instance is faulty. 
 
 
     
     
       18. The computer system of  claim 17 , wherein the computer program instructions further cause the one or more electronic processors to perform steps comprising:
 determining the modified alert threshold value by executing a machine learning based model. 
 
     
     
       19. The computer system of  claim 17 , wherein the modified alert threshold value is determined based on factors including a number of alerts received during a time interval. 
     
     
       20. The computer system of  claim 17 , wherein the modified alert threshold value is determined based on factors including a measure of an indication that an action was taken to adjust the refrigeration equipment instance in response to an alert sent.